Identifying areas of deforestation risk for REDD+ using a species modeling tool

Aguilar-Amuchastegui, Naikoa; Riveros, Juan Carlos; Forrest, Jessica

doi:10.1186/s13021-014-0010-5

Cited by 22 publications

(18 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several machine learning techniques and land cover change prediction models have been developed to project future deforestation based on where it has occurred historically (Harris et al 2008, Fuller et al 2011, Rosa et al 2013, Aguilar-Amuchastegui et al 2014. Despite differences in model structure and complexity, most require at least two land cover maps as well as spatial information about the biophysical and socioeconomic factors that correlate with observed change.…”

Section: Introductionmentioning

confidence: 99%

Using spatial statistics to identify emerging hot spots of forest loss

Harris

Goldman

Gabris³

et al. 2017

Environ. Res. Lett.

185

125

View full text Add to dashboard Cite

As sources of data for global forest monitoring grow larger, more complex and numerous, data analysis and interpretation become critical bottlenecks for effectively using them to inform land use policy discussions. Here in this paper, we present a method that combines big data analytical tools with Emerging Hot Spot Analysis (ArcGIS) to identify statistically significant spatiotemporal trends of forest loss in Brazil, Indonesia and the Democratic Republic of Congo (DRC) between 2000 and 2014. Results indicate that while the overall rate of forest loss in Brazil declined over the 14-year time period, spatiotemporal patterns of loss shifted, with forest loss significantly diminishing within the Amazonian states of Mato Grosso and Rondônia and intensifying within the cerrado biome. In Indonesia, forest loss intensified in Riau province in Sumatra and in Sukamara and West Kotawaringin regencies in Central Kalimantan. Substantial portions of West Kalimantan became new and statistically significant hot spots of forest loss in the years 2013 and 2014. Similarly, vast areas of DRC emerged as significant new hot spots of forest loss, with intensified loss radiating out from city centers such as Beni and Kisangani. While our results focus on identifying significant trends at the national scale, we also demonstrate the scalability of our approach to smaller or larger regions depending on the area of interest and specific research question involved. When combined with other contextual information, these statistical data models can help isolate the most significant clusters of loss occurring over dynamic forest landscapes and provide more coherent guidance for the allocation of resources for forest monitoring and enforcement efforts.

show abstract

Section: Introductionmentioning

confidence: 99%

Using spatial statistics to identify emerging hot spots of forest loss

Harris

Goldman

Gabris³

et al. 2017

Environ. Res. Lett.

185

125

View full text Add to dashboard Cite

show abstract

“…MaxEnt requires presence-only data and information from one point in time to estimate LUCC probabilities. As pointed out by Aguilar-Amuchastegui et al [33], MaxEnt is straightforward to use, free, well tested in a variety of applications, and versatile. Still, while the maximum entropy approach can be used for the identification of areas suitable to experience LUCC, as an SDM, MaxEnt lacks the additional computation step commonly found in standard LUCC models to forecast quantities of change through time (e.g., deforestation rates).…”

Section: Discussionmentioning

confidence: 99%

“…A relatively new approach to model LUCC relies on the Maximum Entropy Principle [28][29][30], extensively adopted for the identification of species niches in ecological studies and presence-only models [31,32]. This approach can be used for the identification of forest areas likely to experience conversion to alternative land uses given a set of environmental and socio-economic restrictions (e.g., [33]). Among the advantages of LUCC models based on the Maximum Entropy Principle is their ability to incorporate multiple data types (e.g., continuous and categorical) and establish complex relationships among predictor variables, through the use of machine-learning algorithms [31].…”

Section: Introductionmentioning

confidence: 99%

Simulated Impacts of Soy and Infrastructure Expansion in the Brazilian Amazon: A Maximum Entropy Approach

Frey

West

Hickler

et al. 2018

Forests

View full text Add to dashboard Cite

Historically, the expansion of soy plantations has been a major driver of land-use/cover change (LUCC) in Brazil. While a series of recent public actions and supply-chain commitments reportedly curbed the replacement of forests by soy, the expansion of the agricultural commodity still poses a considerable threat to the Amazonian and Cerrado biomes. Identification of areas under high risk of soy expansion is thus paramount to assist conservation efforts in the region. We mapped the areas suitable for undergoing transition to soy plantations in the Legal Amazon with a machine-learning approach adopted from the ecological modeling literature. Simulated soy expansion for the year 2014 exhibited favorable validation scores compared to other LUCC models. We then used our model to simulate how potential future infrastructure improvements would affect the 2014 probabilities of soy occurrence in the region. In addition to the 2.3 Mha of planted soy in the Legal Amazon in 2014, our model identified another 14.7 Mha with high probability of soy conversion in the region given the infrastructure conditions at that time. Out of those, pastures and forests represented 9.8 and 0.4 Mha, respectively. Under the new infrastructure scenarios simulated, the Legal Amazonian area under high risk of soy conversion increased by up to 2.1 Mha (14.6%). These changes led to up to 11.4 and 51.4% increases in the high-risk of conversion areas of pastures and forests, respectively. If conversion occurs in the identified high-risk areas, at least 4.8 Pg of CO2 could be released into the atmosphere, a value that represents 10 times the total CO2 emissions of Brazil in 2014. Our results highlight the importance of targeting conservation policies and enforcement actions, including the Soy Moratorium, to mitigate future forest cover loss associated with infrastructure improvements in the region.

show abstract

“…However, limited data availability can hamper their use in some developing countries [20][21][22]. In this paper we present a new approach based on the Random Forest algorithm [23].…”

Section: Introductionmentioning

confidence: 99%

REDD+: Quick Assessment of Deforestation Risk Based on Available Data

Lallo

Mundhenk²,

López³

et al. 2017

Forests

View full text Add to dashboard Cite

Abstract:The evaluation of the future dynamics of deforestation is essential to creating the basis for the effective implementation of REDD+ (Reducing Emissions from Deforestation and forest Degradation) initiatives. Such evaluation is often a challenging task, especially for countries that have to cope with a critical lack of data and capacities, higher uncertainties, and competing interests. We present a new modeling approach that makes use of available and easily accessible data sources to predict the spatial location of future deforestation. This approach is based on the Random Forest algorithm, which is a machine learning technique that enables evidence-based, data-driven decisions and is therefore often used in decision-making processes. Our objective is to provide a straightforward modeling approach that, without requiring cost-intensive assessments, can be applied in the early stages of REDD+, for a stepwise implementation approach of REDD+ projects in regions with limited availability of data, capital, technical infrastructure, or human capacities. The presented model focuses on building business-as-usual scenarios to identify and rank potentially suitable areas for REDD+ interventions. For validation purposes we applied the model to data from Nicaragua.

show abstract

Identifying areas of deforestation risk for REDD+ using a species modeling tool

Cited by 22 publications

References 27 publications

Using spatial statistics to identify emerging hot spots of forest loss

Using spatial statistics to identify emerging hot spots of forest loss

Simulated Impacts of Soy and Infrastructure Expansion in the Brazilian Amazon: A Maximum Entropy Approach

REDD+: Quick Assessment of Deforestation Risk Based on Available Data

Contact Info

Product

Resources

About